Cancer is a major public health problem, being responsible for one in four deaths in Western countries 1. Imaging is used for cancer detection, biological characterization, delineation, and treatment monitoring before and after therapeutic interventions 2. Functional whole-body imaging techniques such as PET play an important role in the evaluation of cancer, complementary to structural imaging techniques 2. Since its introduction in 1976, fluorine-18 fluorodeoxyglucose (18F-FDG) has been and still is the most widely used radiotracer for oncological PET studies 3. In fact, 18F-FDG-PET is used in more than 90% of cancer patients in clinical practice 4,5. Meanwhile, several non-18F-FDG radiotracers are already in use for routine clinical use in oncology, most notably 11C-choline and 18F-choline for prostate cancer, 11C-methionine for brain tumors, 18F-dihydroxyphenylalanine and 68Ga-labeled somatostatin analogues for neuroendocrine tumors, and 11C-acetate for prostate cancer and hepatic tumors 4,5. These non-18F-FDG radiotracers are used because of low or no accumulation of 18F-FDG in several of these cancers and/or the relatively high background 18F-FDG uptake in certain body regions (e.g. brain and urogenital tract), as a result of which these tumors are suboptimally visualized at conventional 18F-FDG-PET (which is usually performed 60 min after an intravenous administration of 18F-FDG) 4,5. Nevertheless, it should be noted that the 18F-FDG avidity of a cancer should not merely be interpreted unidimensionally in terms of tumor detectability, but it also reflects the phenotypic aggressiveness of the tumor, which may have subsequent prognostic implications as has been shown in several cancers 6. In addition, delayed 18F-FDG-PET imaging (i.e. PET imaging obtained more than 60 min after 18F-FDG administration, usually 2 h) can improve the detectability of cancers thanks to the time-dependent increase in 18F-FDG uptake in several cancers and clearance of 18F-FDG activity in several normal tissues (e.g. liver, spleen, and mediastinum) 7,8. Another impetus for the development of non-18F-FDG radiotracers in oncology is the aim to assess tumor physiological or metabolic processes other than glucose metabolism 4,5. Specific targets for PET imaging that are currently under active investigation include cell proliferation imaging, angiogenesis imaging, and oxygenation imaging 2,4,5. Imaging of these biological phenomena is believed to improve tumor detection and characterization, risk stratification, and treatment planning and monitoring. However, the development of new radiotracers is time-consuming, expensive, and burdensome to cancer patients. If these new radiocompounds do not provide any additional information beyond that which can already be obtained by means of the established radiotracer 18F-FDG, then there is no logical reason why further efforts and resources should be devoted to developing these agents. In this scientific communication, the role of specific radiotracers for cell proliferation imaging, angiogenesis imaging, and oxygenation imaging will be critically reviewed and compared with 18F-FDG imaging. Cell proliferation imaging assesses the uptake and metabolism of nucleosides used in nucleic acid synthesis 2. The most widely used radiotracer for cell proliferation imaging is the thymidine analogue 18F-fluorothymidine (FLT), which enters cells by nucleoside transporters (which may be upregulated in cancer cells) and to a lesser extent by passive diffusion. FLT undergoes phosphorylation by thymidine kinase 1 (which is almost absent in quiescent cells, but is increased in proliferating cells, particularly in the S-phase of the cell cycle) to form FLT-monophosphate. FLT-monophosphate may continue to be phosphorylated into FLT-diphosphate and FLT-triphosphate. FLT-triphosphate is not incorporated into DNA, but trapped in the cytosol. Thus, the accumulation of FLT-monophosphate, diphosphate, and triphosphate nucleotides forms the basis for FLT-PET imaging 9,10. Importantly, the halogen substitution in the 3′ position of FLT results in a decreased affinity for the pyrimidine transporter compared with thymidine. Moreover, the affinity of FLT for thymidine kinase 1 has been reported to be 30% less than the affinity of thymidine. Both mechanisms lead to a preferential uptake and phosphorylation of thymidine, which may explain the low FLT uptake in tumor cells compared with thymidine 5. In addition, and importantly, FLT uptake is lower than 18F-FDG uptake in most solid tumors 10, which limits the sensitivity of FLT-PET. Another issue that compromises the sensitivity of this technique is high physiological FLT uptake in the liver and the bone marrow 10, sites that are commonly involved in metastatic disease. Liver FLT uptake has also been reported to obscure adjacent gastrointestinal cancers 11. 18F-FDG-PET is superior to FLT-PET in the evaluation of these anatomical sites that are important in clinical oncology 11–13. Thus, unlike 18F-FDG-PET, FLT-PET imaging has limited value in the systemic assessment of disease activity in the body. Furthermore, although initial studies reported FLT to be a specific tumor tracer 14, FLT is also absorbed by inflammatory lesions associated with reproductive-stage bacteria or viruses, as shown by more recent work 15. For example, FLT is taken up in nonmetastatic reactive lymph nodes because of reactive B-lymphocyte proliferation 16. Moreover, although FLT-PET does not suffer from physiological background brain uptake, it cannot distinguish between benign lesions showing blood–brain barrier disruption and malignant brain tumors 16. It is hoped that FLT-PET could be a potential method for early response assessment 9,10. However, the low sensitivity of FLT-PET is disadvantageous for monitoring lesions during treatment. Furthermore, although most studies have reported a significant early decrease in tumor FLT uptake after therapy, this reduction does not always help to predict response to interventions. In two nonhuman studies evaluating the effects of belinostat on human ovarian cancer cells in mice and cyclophosphamide on Burkitt lymphoma-bearing mice, 18F-FDG-PET was reported to be superior to FLT-PET in detecting early response to therapy 17,18. Detection of response to therapy using FLT also seems to be highly dependent on the administered drugs 17,18 and there are some reports of an increase in tumor FLT uptake after efficient intervention 19,20, which could result from using drugs inhibiting the de-novo pathway of DNA synthesis 21. The decrease in FLT uptake is not always observed with histopathologic response and may be because of arrest induced by therapy rather than cell death 9; these cells may still have the potential to become proliferative. In addition, it has been shown that some extremely proliferative malignant tumors use the de-novo pathway (which does not require external thymidine) instead of the salvage pathway, which results in less FLT uptake, whereas moderately proliferative cells may have much higher FLT uptake because they are using the salvage pathway for DNA synthesis 22. Thus, unlike more general proliferation markers, such as Ki-67 and 18F-FDG, FLT reflects proliferative indices to variable and potentially unreliable extents, and should not be considered a surrogate of the proliferative index 22. Angiogenesis, the growth of new vessels by endothelial cells from pre-existing vessels 23, was introduced as an important mechanism by which a cancer starts to gather enough oxygen and energy for growth. In the absence of angiogenesis, tumors are supposed to be restricted to 2–3 mm in size 24,25. In addition, a primary tumor must gain access to vasculature for distant metastasis and later growth in new regions 26,27. Activation of endothelial cells is often triggered by decreased oxygen and nutrient delivery to tissues 28,29. Hypoxia-inducible factor upregulates vascular endothelial growth factor, matrix metalloproteinase, and many other factors 24,30. This leads to degradation of basement membrane and extracellular matrix and then the next steps of angiogenesis are initiated. Integrins such as αvβ3 are important for adhesion and migration of endothelial cells and also the growth, survival, and differentiation of these cells 30. Many strategies have been introduced to image angiogenesis activity that target vascular endothelial growth factor and its receptor, αvβ3 integrin, fibronectin, and matrix metalloproteinase 27. However, αvβ3 integrin is by far the most studied target for scanning neovascularization in tumors 27. The level of this integrin is very high on neovascularization observed in tumors, but low on epithelial and mature endothelial cells 31. αvβ3 binds to extracellular proteins with the arginine–glycine–aspartate (RGD) sequence. Thus, cyclic RGD peptides have become the core for making tracers targeting this integrin 27,32,33. This tracer has been studied in many human cancers including musculoskeletal, melanoma, colon, breast, head or neck, sarcoma, and osseous metastases 34. Uptake of RGD-based tracers is very heterogeneous in primary tumors and metastasis 25, and this heterogeneity is not associated with tumor size and metabolic activity as measured by 18F-FDG-PET 24–26,34,35. There are relatively high intertumor and intratumor variabilities in tracer uptake in different primary tumor lesions and metastasis that could be because of varying αvβ3 expressions 26 and course of metastatic disease 34. The sensitivity of RGD-PET in detecting malignant lesions is affected by limited expression of its target in many malignancies, low capability to show metastasis, and finally accumulation of tracer in the liver and urinary tract. The target of RGD, αvβ3 integrin, is only expressed in a subset of cancer patients depending on the type and stage of the tumor and may not be very useful 26. Also, this imaging modality has been shown not to be very good at detecting lymph node and distant metastasis of cancers in cases in which both primary tumor and metastasis are known to express αvβ3 24,25, which is probably because of the lack of neovascularization or low levels of αvβ3 in these small lesions. Moreover, accumulation of RGD in the liver 24,26, urinary tract, and bladder 34 is very high and may obscure tumors and metastases with low to moderate angiogenesis activity in and around these organs. The specificity of RGD-PET may be a major problem when physicians work on human diseases. When a patient suspected of having cancer with extensive angiogenesis undergoes neoangiogenesis scanning, it should be considered that, in addition to the growing endothelium, tumor cells themselves 25,31 and also macrophages and smooth muscle cells 26 may express αvβ3 integrin, which would reduce the specificity of RGD as a diagnostic strategy for detecting only growing neovessels. Moreover, angiogenesis is not a specific feature of malignancies only, but it can also be found in inflammatory situations and diseases, tissue remodeling processes, and wound healing 23,36–39. Imaging angiogenesis activity may be useful in a selection of patients to determine the presence of potential therapeutic targets (mostly αvβ3 integrin), prescribe antiangiogenesis and αvβ3 antagonists drugs to prevent lesion progression and metastasis 30,34,40, and then monitor response to these treatments 26,41,42. However, because of the above-mentioned limitations, it is unlikely that angiogenesis PET tracers will either replace or become a practical, widespread adjunct to 18F-FDG. We wish to also point out that hypoxic cells use glucose inefficiently and therefore have high glycolysis, which can be detected by 18F-FDG-PET imaging 43. This is a well-known phenomenon and has been proven over the past decades. This further enhances the importance of 18F-FDG in assessing the degree of hypoxia in these cells 44. Several hypoxic compounds have been introduced during the past two decades and are being proposed as unique markers of hypoxia in various settings 45–48. In particular, such agents are believed to be very useful in patients who are undergoing radiation therapy 49. Hypoxic cells are more resistant to radiation effects than those with normal oxygenation. Therefore, to deliver lethal doses of radiation to hypoxic tissues, larger doses of either internal or external radiation exposure are required for optimal results. Again, we should point out that tissues with significant cell proliferation and hypoxia are associated with angiogenesis 50. Therefore, 18F-FDG-PET imaging will reveal the cumulative effects of these biological parameters as a single imaging modality. Finally, one should note that a decrease or a lack of uptake of 18F-FDG in tumor tissues reflects the degree of aggressiveness of the disease. In other words, tumors that are highly undifferentiated and therefore have significant cell proliferation, hypoxia, and angiogenesis are likely to have high levels of 18F-FDG uptake that can be detected by PET imaging. However, tumors cells that are highly differentiated and therefore are growing slowly will require lower levels of glycolysis to sustain their viability, and thus decreased uptake of 18F-FDG. This phenomenon is observed commonly in tumors that are hormonally driven and known to be less aggressive 6,51. In fact, the majority of successes described with surgical resection have been noted in this category of malignant disorders. The best-characterized tumors in this category are thyroid cancer, breast cancer, prostate cancer, neuroendocrine tumors, and some rare malignancies. These tumors have been investigated by specific tracers that target hormone receptors including estrogen-based compounds in breast cancer and somatostatin-based compounds in neuroendocrine tumors 52. We believe that the use of these specific compounds may also be redundant and may not markedly improve the outcome in these malignancies. Various groups, including our own, have shown that tumors in this category have minimal or no uptake during the first hour after the administration of 18F-FDG, but may become positive over time 53. Therefore, imaging such malignancies at later time points will yield positive results in this population. Thus, 18F-FDG-PET imaging in hormone-driven neoplasia has great prognostic value and predicts the outcome in the majority of patients by a single 18F-FDG-PET scan. Limiting the role of 18F-FDG-PET imaging in these cancers because of perceived false negatives may not be justified. Paradoxically, developments in genetics speak in favor of more generic than very specialized tracers. Thus, recent research – contrary to the traditional model of early dissemination of metastatic cells and parallel progression of primary tumors and metastases – provides evidence of linear progression of breast cancer with relatively late dissemination from the primary tumor and genomic discordance between the different stages of tumor evolution 54,55, meaning that a single, more specific, PET tracer may not be able to identify the full extent and severity of cancer and quantify the entire tumor burden in the body as can be done with 18F-FDG. On the basis of what has been described in this scientific communication, we believe that 18F-FDG-PET as a single imaging modality will remain the dominant tool for assessing a multitude of malignant disorders in the future. Routine use of specific compounds for detecting particular biological processes may often not be justified in the clinical management of many cancers 56. The specific tracers as noted above will play a role only as a research tool for specific purposes and as such will enable better understanding of specific tumor biology and efficacy of future therapeutic interventions. A recent article published in American Journal of Roentgenology describes the obstacles that the imaging community is facing in securing reimbursement from Medicare in the USA 57. In fact, it is becoming clear that even reimbursement for currently approved imaging procedures is decreasing, and therefore, a major effort is being made to reduce clinical utilization of both radiological and nuclear medicine procedures. In this situation, we will be facing significant resistance from government agencies as well as private insurance companies to add newer procedures for routine clinical use of PET. The above considerations may provide the impression that we prefer 18F-FDG over other PET tracers. This does not reflect our intentions in drafting this editorial. Along with our colleagues in the field, we would greatly welcome new tracers that provide advantages over 18F-FDG with great enthusiasm, especially those that have potential clinical significance. Through the years, we have anticipated the development of new tracers that may provide advantages over 18F-FDG for specific clinical applications; however, none have come to fruition at this time. In this respect, it is sad that the excellent Molecular Imaging and Contrast Agent Database is no longer being updated 58. This editorial should be perceived as a call for a fruitful presentation and discussion of promising new tracers that can rival the impact of 18F-FDG while propelling PET/CT and PET/MRI as the modalities of choice for clinical molecular imaging in the 21st century. Acknowledgements Conflicts of interest There are no conflicts of interest.
No takes yet. Share an insight, caveat, or question.
Kwee et al. (2016) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: